plot3Dslice

调用标定好的双目摄像机，利用其内外参数进行立体匹配，计算所拍图像的视差，产生深度图(Call the calibrated binocular camera, use its internal and external parameters for stereo matching, calculate the parallax of the captured image, and generate depth map)

说明：  经典的图像像素层融合算法-Brovery.程序输入中两幅图像需经过配准，大小一致。matlab代码内有详细的注释。其他像素层融合算法可仿照此修改即可。需要配对图像和运行结果、其它基于小波融合代码的可发邮件给我。非常乐意和大家共享编程乐趣。(classic image pixel level fusion algorithm-Brovery. Import procedures which require a two image registration, the same size. Matlab code with detailed comments. Other pixel level fusion algorithm can be modeled on this amendment. Image matching needs and operating results, the other based on wavelet integration code can mail it to me. Very happy to share programming and fun.)

说明：  利用图象边缘特征的基于豪思道夫距离的景象匹配算法，匹配速度要比基于灰度的算法高好几个数量级(use of the images on the edge of Ho Sze-Astoria distance from the scene matching algorithm, speed matching gray than the algorithm based on several high-volume)

红外图像细节增强仿真，其中使用双边滤波器进行图像分层。带测试源数据。(Infrared image detail enhancement, with test source data)

说明：  我们提出了一个简单、灵活和通用的对象实例分割框架。我们的方法能有效检测图像中的对象，同时为每个实例生成高质量的 segmentation mask。这种被称为 Mask R-CNN 的方法通过添加用于预测 object mask 的分支来扩展 Faster R-CNN，该分支与用于边界框识别的现有分支并行。Mask R-CNN 训练简单，只需在以 5fps 运行的 Faster R-CNN 之上增加一个较小的 overhead。此外，Mask R-CNN 很容易推广到其他任务，例如它可以允许同一个框架中进行姿态估计。我们在 COCO 系列挑战的三个轨道任务中均取得了最佳成果，包括实例分割、边界对象检测和人关键点检测。没有任何 tricks，Mask R-CNN 的表现优于所有现有的单一模型取得的成绩，包括 COCO 2016 挑战赛的冠军。(Mask R-CNN code by HeKaiming)

主要包括3种去噪方法，Bayesian Least Squares，BM3d 和Non-local means (this attachment contains three classical denoising methods Bayesian Least Squares，BM3d and Non-local means )

Level Set Methods and Dynamic Implicit Surfaces - Ronald Fedkiw(Level Set Methods and Dynamic Implicit Surfaces- Ronald Fedkiw)

将保存有灰度图像信息的txt文件转化生成灰度图片。 txt数据只能包含数据，数据用逗号或者空格隔开，所存数据个数应与所生成的图片大小相同。例：要还原的图片大小为640*480，那么txt文件中就应当有307200个数据。(The txt file with information of grayscale image is transformed into gray image. The TXT data can only contain data, separated by commas or spaces, and the number of data stored should be the same as the size of the generated image. Example: the size of the image to restore is 640*480, so there should be 307200 data in the txt file.)

说明：  《数字图像处理（第三版）》，中文版，作者是（美）冈萨雷斯，本书是数字图像处理经典著作。("Digital Image Processing (3rd Edition)", Chinese Edition, by Gonzalez, USA. This book is a classic work of digital image processing.)

自己编写的C++小程序，十字算法，用到KD树，能够有效进行点云去噪，压缩数据(I have written a small program C++, cross algorithm used in KD tree, the point cloud can be effectively de-noising, compression of data)